Intelligent positioning and verification method and system for the ovality change part of tunnel segment

Abstract

The present invention relates to the technical field of tunnel ellipticity change positioning and verification methods, in particular to a tunnel segment ellipticity change location intelligent positioning and verification method and system thereof, which obtains location information of multiple two-dimensional code targets on the surface of a tunnel segment; collects multiple groups of three-dimensional point cloud data of the tunnel segment; determines the major axis of a fitted ellipse of the tunnel segment based on the location information of the two-dimensional code targets and the multiple groups of three-dimensional point cloud data; sets multiple scanning points according to the major axis of the fitted ellipse; expands the scanning area with the scanning point as the center and collects detailed three-dimensional point cloud data of the tunnel segment; identifies a deformed location from the detailed three-dimensional point cloud data based on a preset deformation threshold; generates location information and deformation amount data of the deformed location; outputs a safety assessment result based on the location information and deformation amount data of the deformed location, as well as a preset safety threshold, thereby achieving accurate prediction of the long-term deformation trend of the tunnel and providing technical support for formulating preventive maintenance strategies and optimizing tunnel management measures.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel ovality change positioning and verification methods, in particular to an intelligent positioning and verification method and system for a tunnel segment ovality change portion. Background Art

[0002] With the acceleration of urbanization and the increasing use of underground space, tunnel engineering plays an increasingly important role in modern infrastructure construction. However, as a complex underground structure, the long-term safety and stability of tunnels have always been the focus of attention in the engineering community. In particular, the ovality change of tunnel segments, as an important indicator reflecting the health status of tunnel structures, is crucial to ensure the safe operation of tunnels.

[0003] Traditional tunnel segment ovality detection methods mainly rely on manual measurement or simple mechanical equipment. These methods are not only time-consuming and labor-intensive, but also difficult to ensure the accuracy and comprehensiveness of the measurement. With the advancement of technology, 3D laser scanning technology has begun to be applied to the field of tunnel detection, greatly improving the efficiency and accuracy of data collection. However, the existing detection methods based on 3D laser scanning still have some significant shortcomings.

[0004] First, existing methods have difficulties in the registration and processing of point cloud data. The tunnel environment is complex, the lighting conditions are poor, and there is often dirt and water stains on the surface of the segments. These factors will affect the quality of point cloud data and the accuracy of registration. Existing registration algorithms often have difficulty in achieving ideal results when processing data in such a complex environment, which affects the accuracy of subsequent analysis.

[0005] Secondly, existing methods lack intelligence and automation in the identification and location of deformed parts. Most methods still require manual intervention to judge and mark deformed areas, which not only increases the workload but also introduces subjective factors, affecting the objectivity and consistency of the detection results.

[0006] Furthermore, existing methods generally lack an effective verification mechanism for test results. Due to the complexity of tunnel structures and the variability of geological conditions, it is difficult to fully and accurately assess deformation conditions solely based on measurement data. The lack of support from theoretical models and the integration of multi-source data greatly reduces the reliability and persuasiveness of test results.

[0007] In addition, existing methods also have shortcomings in safety assessment and early warning. Most methods use fixed safety thresholds and cannot adapt to the changing characteristics of tunnels at different stages and in different environments. This rigid assessment mechanism may lead to incorrect safety judgments, either being too conservative and causing unnecessary maintenance costs, or being too optimistic and ignoring potential safety hazards.

[0008] Finally, the existing methods also need to be improved in terms of long-term accumulation and utilization of data. The lack of effective analysis and utilization of historical data makes it difficult to accurately predict the long-term deformation trend of the tunnel, which is a major drawback for formulating preventive maintenance strategies and extending the service life of the tunnel.

[0009] In view of the above problems, a new method is urgently needed to accurately locate the parts where the ovality of the tunnel segments changes and reliably verify the test results. This method should be able to overcome the influence of complex environments and achieve high-precision data collection and processing; be able to intelligently identify and locate the deformation parts and reduce human intervention; be able to combine theoretical models and multi-source data to effectively verify the test results; be able to achieve dynamic safety assessment and early warning; and be able to accurately predict the deformation trend of the tunnel based on long-term data accumulation. Summary of the invention

[0010] The present invention is an innovative solution to the above technical problems. By combining advanced 3D laser scanning technology, intelligent algorithms, theoretical models and multi-source data fusion, the present invention provides a comprehensive, accurate and reliable intelligent positioning and verification method for the ovality change of tunnel segments.

[0011] The present invention proposes an intelligent positioning and verification method for a tunnel segment ovality change portion, including:

[0012] The acquisition steps include:

[0013] Obtain the location information of multiple QR code targets on the surface of the tunnel segment;

[0014] Collect multiple sets of 3D point cloud data of tunnel segments;

[0015] Processing steps include:

[0016] Determining the major axis of the fitted ellipse of the tunnel segment based on the position information of the two-dimensional code target and the multiple sets of three-dimensional point cloud data;

[0017] According to the major axis of the fitted ellipse, a plurality of scanning points are set;

[0018] Taking the scanning point as the center, expanding the scanning area and collecting detailed three-dimensional point cloud data of the tunnel segment;

[0019] Based on a preset deformation threshold, identifying a deformed part from the detailed three-dimensional point cloud data;

[0020] Output steps include:

[0021] generating position information and deformation amount data of the deformation part;

[0022] Based on the position information and deformation amount data of the deformation part and a preset safety threshold, a safety assessment result is output.

[0023] Preferably, the obtaining step specifically includes:

[0024] Multiple QR code targets are laid out on the surface of the tunnel segments;

[0025] A three-dimensional laser scanner is used to collect a first set of three-dimensional point cloud data along the tunnel vault direction, taking the two-dimensional code target as a reference point.

[0026] Preferably, determining the major axis of the fitting ellipse of the tunnel segment in the processing step specifically comprises:

[0027] Based on the first set of three-dimensional point cloud data and the pre-established tunnel structure design model, using the SAC-IA algorithm to perform point cloud registration;

[0028] According to the registered point cloud data, the major axis of the fitted ellipse of the tunnel segment is calculated.

[0029] Preferably, setting a plurality of scanning points in the processing step specifically includes:

[0030] In the direction of the major axis of the fitted ellipse, a first scanning point is set at an interval smaller than the accuracy of the segment ovality detection equipment;

[0031] Taking the first scanning point as the center, a scanning line is set every 5° in a direction perpendicular to the major axis of the fitted ellipse;

[0032] On each scanning line, the first scanning point is used as the reference point and extended 2 cm to both ends to form a fan-shaped scanning area.

[0033] Preferably, identifying the deformed part in the processing step specifically includes:

[0034] Calculate the distance between each point cloud data point and the center of the fitted ellipse to obtain the distance index;

[0035] comparing the distance indicator with a preset distance threshold;

[0036] Points whose distance is greater than the distance threshold are marked as abnormal points to form deformation parts.

[0037] Preferably, a verification step is also included:

[0038] Based on the position information of the deformed part, collecting high-precision three-dimensional point cloud data of the part;

[0039] Using an adaptive random sampling consistency algorithm to perform ellipse fitting on the high-precision three-dimensional point cloud data;

[0040] The difference between the fitted ellipse and the designed ellipse is calculated to obtain the precise deformation.

[0041] Preferably, the verification step further comprises:

[0042] Based on the tunnel geological conditions and stress analysis, a three-dimensional deformation prediction model is established;

[0043] The precise deformation amount is compared with the prediction result of the three-dimensional deformation prediction model to verify the rationality of the deformation.

[0044] Preferably, the output step further comprises:

[0045] Calculate the tunnel convergence value of the section where the deformation part is located;

[0046] Comparing the tunnel convergence value with a preset safety threshold;

[0047] When the tunnel convergence value exceeds the safety threshold, an alarm message is generated.

[0048] As a preference, the method further includes a data fusion step:

[0049] Obtain the design parameters and geological condition data of the tunnel segments;

[0050] Based on the design parameters, geological condition data and deformation data, a tunnel segment ovality variation model is established;

[0051] The ellipticity variation model is used to predict the long-term deformation trend of the tunnel segment.

[0052] The intelligent positioning and verification system for the ovality change part of the tunnel segment used to implement the method comprises:

[0053] Data acquisition module for:

[0054] Obtain the location information of multiple QR code targets on the surface of the tunnel segment;

[0055] Collect multiple sets of 3D point cloud data of tunnel segments;

[0056] Data processing module for:

[0057] Determining the major axis of the fitted ellipse of the tunnel segment based on the position information of the two-dimensional code target and the multiple sets of three-dimensional point cloud data;

[0058] According to the major axis of the fitted ellipse, a plurality of scanning points are set;

[0059] Taking the scanning point as the center, expanding the scanning area and collecting detailed three-dimensional point cloud data of the tunnel segment;

[0060] Based on a preset deformation threshold, identifying a deformed part from the detailed three-dimensional point cloud data;

[0061] Result output module, used for:

[0062] generating position information and deformation amount data of the deformation part;

[0063] Outputting a safety assessment result based on the position information and deformation amount data of the deformation part and a preset safety threshold;

[0064] Authentication module, used to:

[0065] Based on the position information of the deformed part, collecting high-precision three-dimensional point cloud data of the part;

[0066] Using an adaptive random sampling consistency algorithm to perform ellipse fitting on the high-precision three-dimensional point cloud data;

[0067] Calculate the difference between the fitted ellipse and the designed ellipse to obtain the precise deformation;

[0068] Early warning module, used to:

[0069] Calculate the tunnel convergence value of the section where the deformation part is located;

[0070] Comparing the tunnel convergence value with a preset safety threshold;

[0071] When the tunnel convergence value exceeds the safety threshold, an alarm message is generated.

[0072] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0073] First, by innovatively using two-dimensional code target technology and efficient point cloud registration algorithm, this method significantly improves the accuracy of data acquisition and processing. Even in complex tunnel environments, sub-millimeter measurement accuracy can be achieved, which lays a solid foundation for subsequent deformation analysis.

[0074] Secondly, this method introduces an intelligent deformation recognition algorithm that can automatically and accurately locate the deformed parts. This not only greatly improves the detection efficiency and reduces human errors, but also can detect tiny deformations that are difficult to detect with the naked eye, making it possible to detect potential problems in a timely manner.

[0075] Furthermore, this method innovatively combines theoretical models and measured data to achieve multi-dimensional verification of the test results. This "theory + practice" verification mechanism greatly improves the reliability and persuasiveness of the test results and provides a more scientific basis for tunnel safety assessment.

[0076] In addition, this method introduces a dynamic safety threshold adjustment mechanism and an intelligent early warning system, which can dynamically adjust the safety assessment criteria according to the actual conditions and historical data of the tunnel, and achieve more accurate risk warnings. This not only improves the efficiency of safety management, but also effectively reduces the risk of false alarms and missed alarms.

[0077] Finally, through multi-source data fusion and machine learning algorithms, this method can accurately predict the long-term deformation trend of the tunnel, which provides strong technical support for formulating preventive maintenance strategies, optimizing tunnel management measures, and extending the service life of the tunnel.

[0078] In general, the method provided by the present invention not only solves many problems existing in the prior art, but also achieves a qualitative leap in accuracy, efficiency, reliability and foresight. Its application will greatly improve the level of tunnel inspection and maintenance, and make an important contribution to ensuring the long-term safe operation of tunnel projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 The figure is a flow chart of the overall method of the present invention.

[0080] Figure 2 It is a detailed logic block diagram of the data acquisition module of the present invention.

[0081] Figure 3 It is a logic block diagram of the data processing module of the present invention.

[0082] Figure 4 It is a logic block diagram of the verification module of the present invention.

[0083] Figure 5 It is a logic block diagram of the early warning module of the present invention.

[0084] Figure 6 It is a logic block diagram of the result output module of the present invention. DETAILED DESCRIPTION

[0085] Please refer to Figure 1-6 The present invention provides a method and system for intelligently locating and verifying the ovality change of a tunnel segment. The method realizes accurate positioning and reliable verification of the ovality change of a tunnel segment by innovatively combining multiple advanced technologies.

[0086] First, the method includes an acquisition step, a processing step, and an output step. In the acquisition step, the position information of multiple two-dimensional code targets on the surface of the tunnel segment is acquired, and multiple groups of three-dimensional point cloud data of the tunnel segment are collected. The advantage of this method is that by arranging multiple two-dimensional code targets on the surface of the tunnel segment, it can not only provide accurate reference points for the three-dimensional laser scanner, but also significantly improve the accuracy of subsequent point cloud registration and deformation analysis.

[0087] In the processing step, the method determines the major axis of the fitting ellipse of the tunnel segment based on the acquired two-dimensional code target position information and multiple sets of three-dimensional point cloud data. The implementation of this step is crucial for the subsequent deformation analysis. Preferably, the present invention adopts the SAC-IA (Sample Consensus Initial Alignment) algorithm for point cloud registration, which is efficient and stable and can achieve high-precision point cloud registration in a complex tunnel environment.

[0088] The core idea of ​​the SAC-IA algorithm is to find the best registration transformation through iteration. Its mathematical expression can be expressed as:

[0089]

[0090] Among them, T * is the optimal transformation matrix, p i and q i are the corresponding points in the source point cloud and the target point cloud respectively, and N is the number of points in the point cloud.

[0091] After determining the major axis of the fitting ellipse, the method sets multiple scanning points according to the major axis. The setting of these scanning points is the key to accurately locate the deformed parts. Specifically, in the direction of the major axis of the fitting ellipse, the first scanning point is set at an interval less than the accuracy of the segment ovality detection equipment. For example, if the accuracy of the detection equipment is 0.1mm, the scanning point interval can be set to 0.05mm to ensure that no potential deformation is missed.

[0092] Next, with the first scanning point as the center, a scanning line is set every 5° in the direction perpendicular to the major axis of the fitted ellipse. This fan-shaped scanning strategy can fully cover the surface of the tunnel segment while maintaining a high scanning efficiency. On each scanning line, the first scanning point is used as the reference point to expand 2 cm at both ends to form a fan-shaped scanning area. This 2 cm expansion distance is determined based on a lot of practical experience and can effectively capture the deformation of the edge of the segment.

[0093] After obtaining detailed 3D point cloud data, this method identifies the deformed parts from the detailed 3D point cloud data based on a preset deformation threshold. The setting of the deformation threshold needs to consider multiple factors, including the design parameters, service life, geological conditions, etc. of the tunnel. In general, the deformation threshold can be set to 0.5%-1% of the thickness of the segment. For example, for a segment with a thickness of 300 mm, the deformation threshold can be set between 1.5 mm and 3 mm.

[0094] The method of the present invention also includes an output step, generating location information and deformation data of the deformed part, and outputting a safety assessment result based on these data and a preset safety threshold. The setting of the safety threshold usually needs to consider factors such as the design service life of the tunnel, load conditions, and geological environment. Generally speaking, the safety threshold can be set to 1.5-2 times the deformation threshold. For example, if the deformation threshold is set to 2mm, the safety threshold can be set between 3mm-4mm.

[0095] In addition, the method of the present invention also includes a verification step, by collecting high-precision three-dimensional point cloud data of the deformed part, and using an adaptive random sampling consistency algorithm to perform ellipse fitting, and calculating the difference between the fitted ellipse and the designed ellipse, thereby obtaining the accurate deformation amount. This multiple verification mechanism greatly improves the reliability of deformation detection.

[0096] The core idea of ​​the adaptive random sampling consensus algorithm is to randomly select sample points from the point cloud data in an iterative manner, fit the ellipse model, and evaluate the quality of the model. Its mathematical expression can be expressed as:

[0097]

[0098] Where E is the error function, p i is a point in the point cloud, f(x i ,y i ) is the fitted elliptic function, and ρ is the robustness loss function.

[0099] Through the above steps, the method of the present invention can accurately locate the deformed parts of the tunnel segments and reliably verify the degree of deformation. This method not only improves the efficiency and accuracy of tunnel detection, but also provides a strong guarantee for the safe operation of the tunnel. The method of the present invention adopts an innovative solution in the acquisition step. Preferably, a plurality of two-dimensional code targets are arranged on the surface of the tunnel segment. These targets not only serve as reference points, but also provide high-precision benchmarks for subsequent point cloud alignment and analysis. The layout of the two-dimensional code targets needs to take into account the curvature of the tunnel and the size of the segment. Usually, 3-5 targets are arranged on each segment to ensure that at least 2-3 targets are in the field of view during scanning.

[0100] In one embodiment of the present invention, a three-dimensional laser scanner is used for data collection. The selection of the scanner needs to comprehensively consider its accuracy, scanning range and speed. For general tunnel inspection, it is recommended to use equipment with an accuracy of not less than 1mm and a scanning range of not less than 30m. When scanning, the two-dimensional code target is used as a reference point and the data is collected along the direction of the tunnel vault. This method can ensure that the complete surface data of the pipe segment is collected while reducing occlusion and blind spots.

[0101] In the data processing step, this method uses the efficient SAC-IA algorithm for point cloud registration. The advantage of this algorithm is that it can quickly process a large amount of point cloud data and maintain high accuracy in the presence of noise and local deformation. The implementation process of the algorithm is as follows:

[0102] 1. Feature extraction: Extract representative feature points from the point cloud, such as corner points, edge points, etc.

[0103] 2. Feature matching: Use a fast nearest neighbor search algorithm to establish preliminary correspondence between the source point cloud and the target point cloud.

[0104] 3. Sampling consistency: Randomly select a set of corresponding points and calculate the transformation matrix.

[0105] 4. Error evaluation: Apply the transformation matrix to the entire point cloud and calculate the registration error.

[0106] 5. Iterative optimization: Repeat steps 3-4 until the best registration result is found or the maximum number of iterations is reached. The mathematical expression of the SAC-IA algorithm can be further refined as follows:

[0107]

[0108] Among them, E(T) is the registration error function, T is the transformation matrix, and p i and q i are the corresponding points in the source point cloud and the target point cloud respectively, w i is the weight coefficient, and N is the number of corresponding points selected.

[0109] Preferably, the method of the present invention also takes into account the geometric characteristics of the tunnel during the registration process. For example, the circular or horseshoe-shaped cross-sectional features of the tunnel can be used to further improve the accuracy of the registration. Specifically, geometric constraints can be added to the registration algorithm:

[0110]

[0111] Among them, C(T) is the geometric constraint function, r j is the distance from the transformed point to the central axis of the tunnel, R is the design radius of the tunnel, and M is the number of selected constraint points.

[0112] By minimizing the weighted sum of E(T) and C(T), the optimal registration result can be obtained that satisfies both the point cloud correspondence and the geometric characteristics of the tunnel.

[0113] After determining the major axis of the fitting ellipse of the tunnel segment, the method adopts a sophisticated scanning point setting strategy. In the direction of the major axis of the fitting ellipse, the interval setting of the scanning points is a key parameter. In a preferred embodiment of the present invention, the interval is set to 1 / 2 to 1 / 3 of the accuracy of the segment ellipticity detection equipment. For example, if the accuracy of the detection equipment is 0.1mm, the scanning point interval can be set to 0.03mm-0.05mm. This setting can ensure that small deformations are not missed while avoiding the generation of too much redundant data.

[0114] In the direction perpendicular to the major axis of the fitted ellipse, this method sets a scan line every 5°. This angle interval is determined based on a lot of practical experience, which can ensure coverage integrity while maintaining high scanning efficiency. It is worth noting that for some special parts, such as seams or known weak areas, the angle interval can be appropriately reduced, for example, adjusted to 2°-3°, to obtain more detailed data.

[0115] On each scan line, the first scan point is used as the reference point to expand 2 cm to both ends to form a fan-shaped scanning area. This 2 cm expansion distance takes into account the possible deformation of the edge of the segment and the special situation of the joint. In practical applications, fine-tuning can be made according to the specific size of the segment and the expected degree of deformation. For example, for larger segments or when larger deformation is expected, the expansion distance can be increased to 2.5-3 cm.

[0116] Through this carefully designed scanning strategy, the method can significantly improve scanning efficiency while ensuring data quality. Compared with the traditional grid scanning method, the method can reduce the scanning time by 30%-50%, while obtaining more targeted and high-quality data.

[0117] In general, the method of the present invention lays a solid foundation for the precise positioning of tunnel segment ellipticity changes by innovatively combining two-dimensional code target technology, efficient point cloud registration algorithm and intelligent scanning strategy. This method not only improves the accuracy and efficiency of detection, but also provides reliable data support for subsequent deformation analysis and safety assessment. The method of the present invention adopts an innovative deformation part identification algorithm in the processing step. The algorithm accurately locates the deformation area based on the distance relationship between the point cloud data and the center of the fitted ellipse. Specifically, the distance between each point cloud data point and the center of the fitted ellipse is first calculated to obtain a distance index. This distance index can be calculated by the following formula:

[0118]

[0119] Among them, D i is the distance index of the i-th point, (x i ,y i) is the coordinate of the i-th point in the point cloud, (x c ,y c ) are the coordinates of the center of the fitted ellipse.

[0120] Preferably, the method adopts an adaptive threshold strategy to identify outliers. Specifically, the distance threshold T can be set as:

[0121] T=μ+kσ

[0122] Among them, μ is the average value of all distance indicators, σ is the standard deviation, and k is the adjustment coefficient. In practical applications, the k value is usually set between 2.5 and 3.5. This adaptive threshold strategy can effectively cope with the differences in deformation characteristics of different tunnel sections.

[0123] In one embodiment of the present invention, in order to further improve the accuracy of recognition, local density analysis is also introduced. By calculating the local neighborhood density of each point, real deformation and noise points can be effectively distinguished. Local density ρ i It can be calculated by the following formula:

[0124]

[0125] Among them, D ij is the distance between the i-th point and the j-th point, d c is the cutoff distance, and N is the total number of points in the point cloud.

[0126] By comprehensively considering the distance index and local density, this method can more accurately identify the deformed parts. For example, a comprehensive scoring function can be defined:

[0127]

[0128] When S i When a certain threshold is exceeded, the point is marked as an abnormal point, thus forming a deformation site.

[0129] In the verification step, the method of the present invention uses high-precision 3D point cloud data acquisition and advanced ellipse fitting algorithm. Preferably, for the identified deformed parts, a higher-resolution 3D laser scanner is used for focused scanning. For example, if the point cloud density of the initial scan is 10 points / cm 2 , then the focus scan on the deformed part can be increased to 50-100 points / cm 2 This refined scan is able to capture even smaller surface variations.

[0130] This method uses an adaptive random sampling consensus algorithm for ellipse fitting. The core idea of ​​the algorithm is to randomly select sample points from point cloud data in an iterative manner, fit the ellipse model, and evaluate the quality of the model. Its mathematical expression can be further refined as:

[0131]

[0132] Where F(x,y)=ax 2 +bxy+cy 2 +dx+ey+f=0 is the general elliptic equation, is the gradient of F, and ρ is the Huber loss function:

[0133]

[0134] Here δ is a threshold parameter, which is usually set to 1.345 to obtain an asymptotic efficiency of 95%.

[0135] The method of the present invention also innovatively introduces a three-dimensional deformation prediction model, combining deformation verification with theoretical analysis. The model is based on tunnel geological conditions and force analysis, and can predict the deformation trend of tunnel segments under different conditions. The core equation of the model can be expressed as:

[0136] ΔR=f(P,E,v,t,r)

[0137] Among them, ΔR is the predicted radial deformation, P is the soil pressure, E is the elastic modulus of the segment material, ν is the Poisson's ratio, t is the segment thickness, and r is the tunnel radius.

[0138] In practical applications, this equation can be further refined as:

[0139]

[0140] Where I is the moment of inertia of the segment, which can be calculated by Calculated.

[0141] By comparing the exact deformation with the results of the prediction model, this method can effectively verify the rationality of the deformation and identify abnormal deformation. For example, if the measured deformation deviates from the predicted value by more than 20%, further investigation and analysis may be required.

[0142] In general, the method of the present invention achieves accurate positioning and reliable verification of changes in the ovality of tunnel segments by innovatively combining high-precision data acquisition, intelligent recognition algorithms and theoretical prediction models. This method not only improves the accuracy and efficiency of detection, but also provides strong technical support for the long-term safe operation of tunnels. The method of the present invention introduces an innovative safety assessment mechanism in the output step. In addition to generating the location information and deformation data of the deformed part, the tunnel convergence value of the section where the deformed part is located is also calculated. This comprehensive assessment method can more comprehensively reflect the safety status of the tunnel.

[0143] Preferably, the tunnel convergence value is calculated using the following formula:

[0144]

[0145] Where C is the convergence rate, D0 is the initial diameter of the tunnel, and D t is the current measured diameter.

[0146] In one embodiment of the present invention, the convergence value calculation also takes into account the irregular deformation of the tunnel. Specifically, multiple measuring points (usually 4-8) are selected on each section, the distance change from each measuring point to the center of the tunnel is calculated, and then the average value is taken as the convergence value of the section. This method can more accurately reflect the overall deformation of the tunnel.

[0147] The setting of the safety threshold is a key link in this method. Usually, the safety threshold is determined comprehensively based on factors such as the design service life of the tunnel, geological conditions, and load conditions. For example, for a tunnel with a design service life of 100 years, the safety threshold in the initial stage (0-5 years) may be set to 0.5%, in the middle stage (5-50 years) may be set to 1%, and in the later stage (50-100 years) may be set to 1.5%.

[0148] This method innovatively introduces a dynamic safety threshold adjustment mechanism. Specifically, the system automatically adjusts the safety threshold based on historical monitoring data and deformation trends. The adjustment formula can be expressed as:

[0149]

[0150] Among them, T new is the new safety threshold, T old is the original safety threshold, k is the adjustment coefficient, is the rate of change of the convergence rate.

[0151] When the tunnel convergence value exceeds the safety threshold, the method will automatically generate an alarm message. The alarm message includes but is not limited to: the exact location of the deformation part, the deformation amount, the convergence value, the degree of exceeding the safety threshold, etc. Preferably, the alarm message also includes deformation trend analysis and preliminary risk assessment.

[0152] The method of the present invention also includes an innovative data fusion step. This step obtains the design parameters and geological condition data of the tunnel segment, and fuses them with the deformation data to establish a tunnel segment ellipticity change model. This multi-source data fusion method can more comprehensively describe and predict the deformation behavior of the tunnel.

[0153] The core equation of the ellipticity change model can be expressed as:

[0154] E(t)=E0+αP(t)+βG(t)+γT(t)+∈

[0155] Among them, E(t) is the ellipticity at time t, E0 is the initial ellipticity, P(t) is the load function, G(t) is the geological condition function, T(t) is the time function, αβ, γ are weight coefficients, and ∈ is the error term.

[0156] This method uses machine learning algorithms, such as support vector regression (SVR) or random forest regression, to train the ellipticity change model. During the model training process, a cross-validation method is used to optimize the model parameters and improve the prediction accuracy.

[0157] Based on the trained ellipticity change model, this method can predict the long-term deformation trend of tunnel segments. This prediction not only takes into account the historical deformation data, but also integrates the influence of design parameters and geological conditions, so it has higher accuracy and reliability.

[0158] Finally, the present invention also provides a tunnel segment ovality change position intelligent positioning and verification system for implementing the above method. The system includes a data acquisition module 1, a data processing module 2, a result output module 3, a verification module 4 and an early warning module 5.

[0159] The data acquisition module 1 is responsible for obtaining the position information of multiple two-dimensional code targets on the surface of the tunnel segment and collecting multiple sets of three-dimensional point cloud data of the tunnel segment. The module includes a two-dimensional code recognition unit 11 and a three-dimensional laser scanning unit 12. The two-dimensional code recognition unit 11 uses a high-resolution camera and advanced image processing algorithms to accurately identify and locate the two-dimensional code targets in a complex tunnel environment. The three-dimensional laser scanning unit 12 uses a high-precision phase laser scanner to quickly obtain high-density point cloud data.

[0160] The data processing module 2 is the core of the system and is responsible for the registration, ellipse fitting and deformation recognition of point cloud data. This module includes a point cloud registration unit 21, an ellipse fitting unit 22 and a deformation recognition unit 23. The point cloud registration unit 21 implements the SAC-IA algorithm described above, the ellipse fitting unit 22 implements the adaptive random sampling consistency algorithm, and the deformation recognition unit 23 implements the outlier recognition algorithm based on distance index and local density.

[0161] The result output module 3 is responsible for generating the position information and deformation data of the deformed parts and outputting the safety assessment results. The module includes a data visualization unit 31 and a report generation unit 32. The data visualization unit 31 can generate an intuitive three-dimensional deformation cloud map, while the report generation unit 32 can automatically generate a test report containing detailed deformation data and safety assessment results.

[0162] The verification module 4 is responsible for further verification of the identified deformed parts. The module includes a high-precision scanning unit 41 and a theoretical analysis unit 42. The high-precision scanning unit 41 uses a higher-resolution three-dimensional laser scanner to focus on scanning the deformed parts, while the theoretical analysis unit 42 implements the three-dimensional deformation prediction model mentioned above.

[0163] The early warning module 5 is responsible for calculating the tunnel convergence value, comparing the safety threshold, and generating alarm information when necessary. The module includes a convergence calculation unit 51, a threshold comparison unit 52, and an alarm generation unit 53. The convergence calculation unit 51 implements the convergence value calculation method described above, the threshold comparison unit 52 implements the dynamic safety threshold adjustment mechanism, and the alarm generation unit 53 is responsible for generating detailed alarm information.

[0164] Through this modular system design, the present invention realizes the full-process intelligent monitoring and analysis of tunnel segment ovality changes. The system not only improves the efficiency and accuracy of detection, but also provides strong technical support for the long-term safe operation of the tunnel.

[0165] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for intelligently locating and verifying the location of the tunnel segment ovality change, characterized in that: include: The acquisition steps include: Obtain the location information of multiple QR code targets on the surface of the tunnel segment; Collect multiple sets of 3D point cloud data of tunnel segments; Processing steps include: Determining the major axis of the fitted ellipse of the tunnel segment based on the position information of the two-dimensional code target and the multiple sets of three-dimensional point cloud data; According to the major axis of the fitted ellipse, a plurality of scanning points are set; Taking the scanning point as the center, expanding the scanning area and collecting detailed three-dimensional point cloud data of the tunnel segment; Based on a preset deformation threshold, identifying a deformed part from the detailed three-dimensional point cloud data; Output steps include: generating position information and deformation amount data of the deformation part; Based on the position information and deformation amount data of the deformation part and a preset safety threshold, a safety assessment result is output.

2. The method according to claim 1, characterized in that: The acquisition step specifically includes: Multiple QR code targets are laid out on the surface of the tunnel segments; A three-dimensional laser scanner is used to collect a first set of three-dimensional point cloud data along the tunnel vault direction, taking the two-dimensional code target as a reference point.

3. The method according to claim 1, characterized in that Determining the major axis of the fitting ellipse of the tunnel segment in the processing step specifically includes: Based on the first set of three-dimensional point cloud data and the pre-established tunnel structure design model, using the SAC-IA algorithm to perform point cloud registration; According to the registered point cloud data, the major axis of the fitted ellipse of the tunnel segment is calculated.

4. The method according to claim 1, characterized in that: The processing step of setting a plurality of scanning points specifically includes: In the direction of the major axis of the fitted ellipse, a first scanning point is set at an interval smaller than the accuracy of the segment ovality detection equipment; Taking the first scanning point as the center, a scanning line is set every 5° in a direction perpendicular to the major axis of the fitted ellipse; On each scanning line, the first scanning point is used as the reference point and extended 2 cm to both ends to form a fan-shaped scanning area.

5. The method according to claim 1, characterized in that The processing step of identifying the deformed part specifically includes: Calculate the distance between each point cloud data point and the center of the fitted ellipse to obtain the distance index; comparing the distance indicator with a preset distance threshold; Points whose distance is greater than the distance threshold are marked as abnormal points to form deformation parts.

6. The method according to claim 1, characterized in that Also includes a verification step: Based on the position information of the deformed part, collecting high-precision three-dimensional point cloud data of the part; Using an adaptive random sampling consistency algorithm to perform ellipse fitting on the high-precision three-dimensional point cloud data; The difference between the fitted ellipse and the designed ellipse is calculated to obtain the precise deformation.

7. The method according to claim 6, characterized in that The verification step further comprises: Based on the tunnel geological conditions and stress analysis, a three-dimensional deformation prediction model is established; The precise deformation amount is compared with the prediction result of the three-dimensional deformation prediction model to verify the rationality of the deformation.

8. The method according to claim 1, characterized in that The output step further comprises: Calculate the tunnel convergence value of the section where the deformation part is located; Comparing the tunnel convergence value with a preset safety threshold; When the tunnel convergence value exceeds the safety threshold, an alarm message is generated.

9. The method according to claim 1, characterized in that: It also includes the data fusion step: Obtain the design parameters and geological condition data of the tunnel segments; Based on the design parameters, geological condition data and deformation data, a tunnel segment ovality variation model is established; The ellipticity variation model is used to predict the long-term deformation trend of the tunnel segment.

10. A system for intelligently locating and verifying the ovality change position of a tunnel segment for implementing the method according to any one of claims 1 to 9, characterized in that: include: Data acquisition module for: Obtain the location information of multiple QR code targets on the surface of the tunnel segment; Collect multiple sets of 3D point cloud data of tunnel segments; Data processing module for: Determining the major axis of the fitted ellipse of the tunnel segment based on the position information of the two-dimensional code target and the multiple sets of three-dimensional point cloud data; According to the major axis of the fitted ellipse, a plurality of scanning points are set; Taking the scanning point as the center, expanding the scanning area and collecting detailed three-dimensional point cloud data of the tunnel segment; Based on a preset deformation threshold, identifying a deformed part from the detailed three-dimensional point cloud data; Result output module, used for: generating position information and deformation amount data of the deformation part; Outputting a safety assessment result based on the position information and deformation amount data of the deformation part and a preset safety threshold; Authentication module for: Based on the position information of the deformed part, collecting high-precision three-dimensional point cloud data of the part; Using an adaptive random sampling consistency algorithm to perform ellipse fitting on the high-precision three-dimensional point cloud data; Calculate the difference between the fitted ellipse and the designed ellipse to obtain the precise deformation; the early warning module is used to: Calculate the tunnel convergence value of the section where the deformation part is located; Comparing the tunnel convergence value with a preset safety threshold; When the tunnel convergence value exceeds the safety threshold, an alarm message is generated.

Images